CN117612619A - Intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst - Google Patents
Intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst Download PDFInfo
- Publication number
- CN117612619A CN117612619A CN202311595841.7A CN202311595841A CN117612619A CN 117612619 A CN117612619 A CN 117612619A CN 202311595841 A CN202311595841 A CN 202311595841A CN 117612619 A CN117612619 A CN 117612619A
- Authority
- CN
- China
- Prior art keywords
- data
- gas
- model
- fischer
- flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000003054 catalyst Substances 0.000 title claims abstract description 93
- 238000003786 synthesis reaction Methods 0.000 title claims abstract description 68
- 238000007726 management method Methods 0.000 title claims abstract description 63
- 230000015572 biosynthetic process Effects 0.000 title claims abstract description 43
- 239000002002 slurry Substances 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 claims abstract description 79
- 230000008569 process Effects 0.000 claims abstract description 46
- 238000004458 analytical method Methods 0.000 claims abstract description 44
- 238000013499 data model Methods 0.000 claims abstract description 39
- 230000036541 health Effects 0.000 claims abstract description 33
- 238000011156 evaluation Methods 0.000 claims abstract description 22
- 230000007246 mechanism Effects 0.000 claims abstract description 14
- 239000007789 gas Substances 0.000 claims description 262
- 239000003921 oil Substances 0.000 claims description 54
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 45
- 238000006243 chemical reaction Methods 0.000 claims description 43
- 238000005457 optimization Methods 0.000 claims description 31
- 239000002994 raw material Substances 0.000 claims description 26
- 239000001257 hydrogen Substances 0.000 claims description 24
- 229910052739 hydrogen Inorganic materials 0.000 claims description 24
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 claims description 22
- 238000006477 desulfuration reaction Methods 0.000 claims description 17
- 230000023556 desulfurization Effects 0.000 claims description 17
- 239000007788 liquid Substances 0.000 claims description 17
- 238000007664 blowing Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 239000000295 fuel oil Substances 0.000 claims description 15
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 claims description 9
- 230000001364 causal effect Effects 0.000 claims description 9
- 230000008447 perception Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000000638 solvent extraction Methods 0.000 claims description 9
- 229910052717 sulfur Inorganic materials 0.000 claims description 9
- 239000011593 sulfur Substances 0.000 claims description 9
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 claims description 8
- 238000010219 correlation analysis Methods 0.000 claims description 7
- 238000010801 machine learning Methods 0.000 claims description 7
- OKKJLVBELUTLKV-UHFFFAOYSA-N Methanol Chemical compound OC OKKJLVBELUTLKV-UHFFFAOYSA-N 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 6
- 230000000087 stabilizing effect Effects 0.000 claims description 6
- 238000000513 principal component analysis Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 4
- 238000013461 design Methods 0.000 claims description 4
- 238000005265 energy consumption Methods 0.000 claims description 4
- 239000000377 silicon dioxide Substances 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 3
- 150000001336 alkenes Chemical class 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000013523 data management Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 150000002431 hydrogen Chemical class 0.000 claims description 3
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- JRZJOMJEPLMPRA-UHFFFAOYSA-N olefin Natural products CCCCCCCC=C JRZJOMJEPLMPRA-UHFFFAOYSA-N 0.000 claims description 3
- 238000011112 process operation Methods 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 235000012239 silicon dioxide Nutrition 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 239000002699 waste material Substances 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000007670 refining Methods 0.000 claims description 2
- 238000005192 partition Methods 0.000 claims 2
- 229910052799 carbon Inorganic materials 0.000 abstract description 5
- 238000004088 simulation Methods 0.000 abstract description 3
- 239000001993 wax Substances 0.000 description 37
- 238000000746 purification Methods 0.000 description 18
- 238000006722 reduction reaction Methods 0.000 description 18
- 238000005262 decarbonization Methods 0.000 description 13
- 230000009467 reduction Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 10
- 230000008859 change Effects 0.000 description 8
- 239000012071 phase Substances 0.000 description 8
- 238000004519 manufacturing process Methods 0.000 description 7
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 6
- 238000005261 decarburization Methods 0.000 description 6
- 150000002430 hydrocarbons Chemical class 0.000 description 6
- 239000004215 Carbon black (E152) Substances 0.000 description 5
- 238000007621 cluster analysis Methods 0.000 description 5
- 229930195733 hydrocarbon Natural products 0.000 description 5
- 238000000926 separation method Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000013145 classification model Methods 0.000 description 4
- 239000007791 liquid phase Substances 0.000 description 4
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 3
- 229910002092 carbon dioxide Inorganic materials 0.000 description 3
- 239000001569 carbon dioxide Substances 0.000 description 3
- 229910002091 carbon monoxide Inorganic materials 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 3
- 238000006555 catalytic reaction Methods 0.000 description 3
- 238000001816 cooling Methods 0.000 description 3
- 238000006731 degradation reaction Methods 0.000 description 3
- 238000013433 optimization analysis Methods 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 206010040844 Skin exfoliation Diseases 0.000 description 2
- 230000035618 desquamation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 239000012188 paraffin wax Substances 0.000 description 2
- 238000005191 phase separation Methods 0.000 description 2
- 238000005498 polishing Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 150000001298 alcohols Chemical class 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 150000001735 carboxylic acids Chemical class 0.000 description 1
- 230000003197 catalytic effect Effects 0.000 description 1
- 239000007809 chemical reaction catalyst Substances 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006392 deoxygenation reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 229910001873 dinitrogen Inorganic materials 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 239000006260 foam Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 125000004435 hydrogen atom Chemical group [H]* 0.000 description 1
- VUZPPFZMUPKLLV-UHFFFAOYSA-N methane;hydrate Chemical compound C.O VUZPPFZMUPKLLV-UHFFFAOYSA-N 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
- 238000009834 vaporization Methods 0.000 description 1
- 230000008016 vaporization Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/10—Analysis or design of chemical reactions, syntheses or processes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
Abstract
The invention provides an intelligent operation management method of a Fischer-Tropsch synthesis slurry bed catalyst, which comprises the following steps: collecting historical operation data in the Fischer-Tropsch synthesis process, and carrying out data treatment to obtain treated data samples; establishing a mechanism model and a process model; data characteristic analysis, namely, finding out the relation among different types of data; dividing working conditions, and constructing and outputting a data model; according to the data model simulation, drawing a parameter dynamic health datum line applicable to different working conditions; the model evaluation is carried out, and the model evaluation is deployed as a functional module component after being qualified; and collecting real-time operation data in the Fischer-Tropsch synthesis process, and comparing the established functional module with the output result of a data model in the functional module. The invention can make the Fischer-Tropsch synthesis slurry bed catalyst run with high performance, improves the performance of the overall Fischer-Tropsch synthesis reaction, solves the problems of stability of the variable outlet product and reasonable configuration of the hydrogen-carbon ratio, and ensures the health and stability of the shift catalyst in the Fischer-Tropsch synthesis reaction process.
Description
Technical Field
The invention relates to the technical field of auxiliary operation in the chemical industry, in particular to an intelligent operation management method for a Fischer-Tropsch synthesis slurry bed catalyst.
Background
The Fischer-Tropsch synthesis reaction uses hydrogen and carbon monoxide as raw materials, and the raw materials are contacted with catalyst particles distributed in a liquid phase in a slurry bed at proper temperature and pressure through a certain proportion, and the chemical technology for synthesizing hydrocarbon is realized through the catalytic action of the catalyst. Hydrocarbon products are all C1-C80, and during the process, a large amount of synthetic water and a small amount of oxygen-containing organic matters such as mixed alcohols, carboxylic acids and the like are generated. In the Fischer-Tropsch synthesis reaction process, the carbon monoxide conversion rate, the selectivity and yield of products, such as the selectivity and yield of high-carbon hydrocarbon, light oil, heavy oil and heavy wax, the ratio of olefin or straight-chain hydrocarbon, the oil-water ratio, the unit consumption of raw gas, the unit consumption of catalyst and other core performance indexes, can finally produce different results according to different compositions of hydrogen and carbon monoxide, sulfur content, hydrogen-carbon ratio and load in the raw gas, different catalyst states, different reserves and different performance, and different control schemes of reaction operation parameters. The performance of the Fischer-Tropsch synthesis reaction is directly affected by the performance state of the catalyst.
The reduced activity of the catalyst accelerates the performance degradation of the fischer-tropsch catalyst. In the operation process, because of the large number of influencing factors and control variables and the complexity of the controlling factors, the Fischer-Tropsch synthesis catalyst is difficult to stably operate in a better interval, and the composition and the property of the product after the catalytic reaction, the yield of the target product, the raw material consumption and the energy consumption in the reaction process, such as raw material gas unit consumption, product yield, compressor circulation amount, carbon dioxide emission and the like, are directly influenced. The complex and lack of accurate sensing operation process results in insufficient performance of the catalyst, difficult accurate control of reaction direction and depth, and difficult optimization of product yield and quality and production performance.
However, the efficiency and energy efficiency of the Fischer-Tropsch synthesis catalytic reaction are determined by the operation performance of the catalyst, and in the prior art, rapid degradation and damage of the Fischer-Tropsch synthesis slurry bed catalyst occur, namely, the lack of effective perception of the performance of the catalyst in the use process is a specific reason; and secondly, the effective basis for the use and maintenance of the catalyst is lacking. The main pain points of the Fischer-Tropsch synthesis catalytic reaction section comprise no catalyst situation and detail perception in the management, so that the production running situation is difficult to be preoccupied and the early warning is triggered; the health state of the catalyst and the lack of real-time analysis and evaluation of the running environment; lack of means for operational life and life cycle prediction; correlation analysis of physical property data, process parameters and performance indexes is lacking; the working condition optimizing means is not available, and a virtuous circle between production decision and production control is difficult to establish; the metadata quality is uneven, the metadata cannot be quickly applied, and the data is changed; the data value is not deeply mined and exploited.
Disclosure of Invention
Aiming at the defects existing in the prior art, the invention aims to provide an intelligent operation management method for a Fischer-Tropsch synthesis slurry bed catalyst, so as to solve the problems in the background art, enable the Fischer-Tropsch synthesis slurry bed catalyst to operate with high performance, improve the performance of the overall Fischer-Tropsch synthesis reaction, solve the problems of stability of a variable outlet product and reasonable configuration of hydrogen-carbon ratio, and ensure the health and stability of a conversion catalyst in the Fischer-Tropsch synthesis reaction process.
In order to achieve the above object, the present invention is realized by the following technical scheme: an intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst comprises the following steps:
collecting DCS, SCADA, PLC in the Fischer-Tropsch synthesis reaction process and historical operation data of an application system, wherein the historical operation data comprise working condition parameters, working condition parameters and target data, the working condition parameters comprise raw material gas composition, raw material gas H/C, raw material gas sulfur content, raw material gas feeding amount and raw material gas hydrogen distribution amount, and carrying out data treatment on the data to obtain treated data samples;
step two, a mechanism model and a process model are built by using the sample data in the step one, and the process model is built according to a Fischer-Tropsch synthesis reaction design file, a DCS picture and operation habits, so that the operation process and related data of the reaction process are simulated;
thirdly, carrying out data characteristic analysis by utilizing the output data of the mechanism model, the output data of the process model and the data sample in the first step, and finding out the relation between different types of data;
dividing working conditions by utilizing data characteristics in the third step, wherein the flow rate of raw materials at the inlet of the reactor and CO in the crude synthesis gas in the third step 2 、H 2 S content, CH 4 After the data of different areas of the feed gas H/C and the feed gas hydrogen distribution data are combined, forming a plurality of working condition areas, and constructing a data model based on different operation data sets in the working condition areas;
step five, evaluating the practicability, applicability, accuracy, errors and effectiveness of the data model obtained in the step four and the feasibility of the model, deploying the data model as a functional module assembly after the data model is qualified, and returning the unqualified data model to the step one for processing due to the problem of data samples; returning the unqualified data model to the second step for treatment due to the mechanism model problem; returning the unqualified data model to the third step for processing due to the characteristic analysis problem;
step six, collecting real-time operation data in the reaction process of preparing olefin from methanol, accessing the functional module established in the step five, and comparing the data with the output result of a data model in the functional module; pushing an optimal operation control scheme under the current working condition when the result index or the performance index is degraded according to the comparison result pushing operation suggestion; and when the result index or the performance index is kept stable, namely the deviation from the output result value is within a certain range, the current situation is kept, and the index optimization operation is not performed.
Further, the method comprises the steps of, the operation data in the first step comprises raw gas component analysis data, tower gas component analysis data, circulating gas component analysis data, decarbonizing component analysis data, raw gas flow, tower gas flow, circulating gas flow, decarbonizing flow, low-alpha tail gas total sulfur, circulating gas component of a self-circulating gas compressor, tower gas flow, decarbonizing and purifying gas flow of a tail gas decarbonizing unit, medium-temperature synthetic water flow, outlet light oil flow, heavy wax pipeline flow, steam drum outlet steam flow, outlet steam flow indication, outlet circulating water flow, fischer-Tropsch and purifying gas flow of a self-refining desulfurization unit, hydrogen gas flow of a tail gas conversion device PSA unit, outlet heavy oil flow, turbine steam inlet flow of a circulating gas compressor, back blowing gas flow of a tail gas decarbonizing unit deoxygenation water flow, self-refined desulfurization unit Fischer-Tropsch purification gas flow, heavy oil liquid level, light oil water level, heavy wax liquid level, inlet high-temperature oil gas and circulating gas differential pressure, inner demister upper and lower low-temperature oil gas differential pressure, self-refined desulfurization unit Fischer-Tropsch purification gas pressure, self-tail gas decarbonization unit decarbonization purification gas pressure, self-tail gas conversion device PSA unit hydrogen pressure, light oil pressure, inlet circulating gas pressure, cold side differential pressure, outlet circulating gas pressure, top low-temperature oil gas pressure, top release gas pressure, wax/oil/gas all temperatures, reactor bottom spent catalyst slurry all temperatures, reactor heavy wax temperature, catalyst slurry temperature, reactor top high-temperature oil gas temperature, self-refined desulfurization unit Fischer-Tropsch purification gas temperature, self-tail gas decarbonization unit decarbonization purification gas temperature, self-tail gas conversion device PSA unit hydrogen pressure, inlet circulating gas pressure, top low-temperature, reactor bottom spent catalyst slurry all temperatures, reactor heavy wax/oil/gas temperatures, catalyst slurry all temperatures, reactor top high-temperature of the self-tail gas conversion device PSA unit, the method comprises the steps of deviation of the temperature of an outlet of a reactor, outlet temperature of a cooler of the outlet of the reactor, circulating gas component of a self-circulating gas compressor, circulating gas component of the self-circulating gas compressor, silicon dioxide, conductivity, flow rate of gas entering a tower, decarbonizing and purifying gas flow rate of a tail gas decarbonizing unit, flow rate of hydrogen of a PSA unit of a tail gas conversion device, indication of middle-temperature synthetic water flow rate, flow rate of stabilizing heavy wax to a stripping tower, flow rate of a heavy wax pipeline, flow rate of the heavy wax pipeline, flow rate of turbine steam inlet of the circulating gas compressor, back-blowing gas flow rate of a tail gas decarbonizing unit, back-blowing gas flow rate of the tail gas decarbonizing unit and oil-water boundary position.
Further, the target data comprise light oil yield, heavy wax yield, synthetic water yield, CO conversion rate, selectivity of each product, catalyst activity index, product yield, catalyst consumption, unit energy consumption and unit consumption of raw material gas.
Further, the data management in the first step includes one or more of the following management methods: data cleaning, data filtering, data setting, data standardization and data alignment.
Further, the data feature analysis in the third step includes one or more of the following analysis methods:
correlation analysis, namely, finding the size and positive-negative correlation of correlation coefficients among different types of data;
principal component analysis, finding out main influencing factors of data;
causal link analysis, namely, finding causal relation of influencing factors of data;
and (3) an expert experience analysis method, which integrates field expert experiences and finds out an empirical optimization rule boundary.
Further, the data model includes.
Further, in the fourth step, dynamic health datum lines applicable to different working conditions and corresponding index health value ranges are simulated and drawn according to the data model; in the health parameter range, finding out the optimal operation parameters and schemes corresponding to the optimal performance indexes under different working conditions through an optimizing model; the data model comprises one or more of a predictive analysis model, a steady-state operation model, a collaborative optimization model and a performance management model.
Further, a predictive analysis model is used for predicting the service life of the catalyst based on machine learning, and indexes such as CO conversion rate, dry gas, light oil, heavy wax selectivity, oil-water ratio, use duration, catalyst storage, unit consumption and the like are used as target indexes, a regression model is built, a machine learning mechanism is built, and a supervised algorithm is adopted to predict the residual service life of the catalyst;
the steady-state operation model is used for constructing an advanced early-warning model based on the health early-warning management applied by the advanced early-warning model, and selecting data of normal working conditions of the catalyst to realize the prediction of health values; based on an approximation degree evaluation algorithm, evaluating the difference degree between an actual value and a predicted value, and realizing early warning of index abnormality; the purposes of interfering the process operation in advance and ensuring the long-term healthy operation of the catalyst are achieved;
the collaborative optimization model is constructed based on an optimal process sample library of a partitioning algorithm, the partitioning algorithm is adopted to perform on-site working condition clustering, the working conditions are partitioned, a working condition identification model is built according to working condition partitioning results, and optimal process parameters of each working condition are calculated;
the performance management model combines the self-defined health threshold value and the optimization range threshold value of the service requirement through data acquisition of each module of the system, combines the management of the thresholds to form performance evaluation indexes of practice and operation processes, ranks the performance of each management object based on the performance evaluation indexes, comprehensively carries out performance evaluation and management according to different time granularity, and forms a closed-loop management model of services such as perception, alarm, intervention, health guarantee, working condition optimizing, performance evaluation, prediction and the like.
Further, the construction method of the data model comprises one or more of the following methods: neural network method, support vector machine method, decision tree method, multiple linear regression method, and random forest.
Further, the evaluation method in the fifth step includes one or more of the following methods: model accuracy method, root mean square error method, root mean square logarithmic error method, and relative error method.
The invention has the beneficial effects that:
1. the intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst can enable the Fischer-Tropsch synthesis slurry bed catalyst to operate with high performance, improve the performance of the overall Fischer-Tropsch synthesis reaction, solve the problems of stability of a variable outlet product and reasonable configuration of the hydrogen-carbon ratio, and ensure the health and stability of the conversion catalyst in the Fischer-Tropsch synthesis reaction process.
2. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst is used for realizing the adjustment of the catalyst under different subdivision working conditions, the upstream parameter change is adapted through operation, so that the catalyst is ensured to keep high-performance operation, the adopted optimizing boundary parameter is based on the parameter of health management, the change and adjustment of the working conditions are automatically matched, and the operation scheme adapted to the current working condition is pushed; the system realizes the functions of performance management and auxiliary operation and maintenance of auxiliary operation, predictively adjusts the auxiliary working condition, keeps high-performance operation, realizes auxiliary process indexes, achieves the operation and lean management.
3. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst realizes real-time online steady state management and health management of the catalyst, ensures the stable operation of the catalyst, and reduces the damage and degradation probability; the method can realize real-time digital perception of catalyst management, such as perception of performance change, perception of change trend, prediction of future change, and perception of details and situation in management.
Drawings
FIG. 1 is a schematic diagram of a data management flow in an embodiment of the invention;
FIG. 2 is a schematic diagram of a correlation analysis flow in an embodiment of the invention;
FIG. 3 is a schematic diagram of a principal component analysis flow in an embodiment of the invention;
FIG. 4 is a schematic diagram of a causal link analysis process in accordance with an embodiment of the present invention;
FIG. 5 is a schematic diagram of a process for analyzing an operational health index according to the present invention;
FIG. 6 is a schematic diagram of a steady-state optimization flow based on health values in an embodiment of the present invention;
FIG. 7 is a schematic diagram of a model deployment flow in an embodiment of the present invention;
FIG. 8 is a schematic flow chart of a system management model according to an embodiment of the present invention;
FIG. 9 is a visual interface diagram of a Fischer-Tropsch synthesis reaction catalyst management system according to an embodiment of the present invention;
FIG. 10 is a flow chart of the overall system solution according to the present invention.
Detailed Description
The invention is further described in connection with the following detailed description, in order to make the technical means, the creation characteristics, the achievement of the purpose and the effect of the invention easy to understand.
Referring to fig. 1 to 10, the present invention provides a technical solution: an intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst has an overall method flow shown in FIG. 10; the management method comprises the steps of,
s1, collecting DCS, SCADA, PLC and historical operation data of an application system in the Fischer-Tropsch synthesis reaction process, and carrying out data treatment by adopting the treatment method: data cleaning, data filtering, data setting, data standardization and data alignment; performing abnormal value processing, data fusion and standardization of data by utilizing data preprocessing, wherein the data is subjected to the technologies of cleaning, setting, alignment, standardization and the like to obtain a treated data sample, and the processing method is shown in figure 1; the historical operating data includes:
working condition parameters: feed gas composition, feed gas H/C, feed gas sulfur content, feed gas feed amount, feed gas hydrogen preparation amount;
the operating data, i.e. the controlled and control parameters, the method comprises raw gas component analysis data, tower entering gas component analysis data, circulating gas component analysis data, decarbonizing component analysis data, raw gas flow, tower entering gas flow, circulating gas component flow, decarbonizing flow, low-grade tail gas total sulfur, circulating gas component of a self-circulating gas compressor, tower entering gas flow, decarbonizing and purifying gas flow of a self-tail gas decarbonizing unit, medium-temperature synthetic water flow, outlet light oil flow, heavy wax pipeline flow, steam drum outlet steam flow, outlet steam flow indication, outlet circulating water flow, self-refined desulfurization unit Fischer-Tropsch purified gas flow, self-tail gas conversion device PSA unit hydrogen flow, outlet heavy oil flow, circulating gas compressor turbine steam inlet flow, back blowing gas flow of a tail gas decarbonizing unit, deoxidizing water flow, self-refined desulfurization unit Fischer-Tropsch purified gas flow heavy oil liquid level, light oil liquid level, heavy wax liquid level, inlet high-temperature oil gas and circulating gas differential pressure, upper and lower low-temperature oil gas differential pressure of an internal demister, fischer-Tropsch purification gas pressure of a self-refined desulfurization unit, decarbonization purification gas pressure of a self-tail gas decarbonization unit, hydrogen pressure of a PSA unit of a self-tail gas conversion device, light oil pressure, inlet circulating gas pressure, cold side differential pressure, outlet circulating gas pressure, top low-temperature oil gas pressure, top release gas pressure, wax/oil/gas all temperatures, all temperatures of waste catalyst slurry at the bottom of a reactor, heavy wax temperature of the reactor, catalyst slurry temperature, top high-temperature oil gas temperature of the reactor, fischer-Tropsch purification gas temperature of the self-refined desulfurization unit, decarbonization purification gas temperature of the self-tail gas decarbonization unit, hydrogen temperature of the PSA unit of the self-tail gas conversion device, temperature deviation of the outlet of the reactor, the outlet temperature of the reactor outlet cooler, the circulating gas component of the self-circulating gas compressor, silicon dioxide, conductivity, tower inlet gas flow, decarbonizing and purifying gas flow of the tail gas decarbonizing unit, hydrogen flow of the PSA unit of the tail gas conversion device, middle-temperature synthetic water flow indication, stable heavy wax to the stripper, heavy wax pipeline flow, turbine steam inlet flow of the circulating gas compressor, back blowing gas flow of the tail gas decarbonizing unit and oil-water boundary position;
Target data: i.e., result parameters including light oil yield, heavy wax yield, synthetic water yield, CO conversion, selectivity of each product, catalyst activity index, product yield, catalyst consumption, unit energy consumption, unit consumption of feed gas.
The Fischer-Tropsch synthesis flow is described:
the Fischer-Tropsch synthesis device for oil products is divided into a synthesis reaction unit, a catalyst reduction unit and a fractional stripping and release gas compression part. The main application and research scenarios are a Fischer-Tropsch synthesis reaction unit and an FT catalyst reduction unit, and the main data sources are the flow and component data of the two devices and other devices.
FT synthesis reaction unit
The FT synthesis reaction flow is that F-T purified synthesis gas with total sulfur content less than 0.05ppm from a fine desulfurization unit is mixed with recycle gas from a recycle gas compressor, recycle hydrogen from a PSA unit and decarbonized recycle gas from a tail gas decarbonizing unit, and the mixed gas enters a recycle heat exchange separator to exchange heat with high-temperature oil gas from the top of a reactor and then enters a Fischer-Tropsch synthesis reactor. The synthesis gas entering the reactor is bubbled through a slurry bed containing the catalyst in the form of bubbles through a gas distributor at the bottom of the reactor to effect the fischer-tropsch synthesis reaction. The light hydrocarbon compound, carbon dioxide, water and unreacted synthesis gas generated by the reaction are led out from the top of the reactor in a gas phase form, and the heavy hydrocarbon generated by the reaction is discharged from the middle of the reactor to a heavy wax collecting tank as heavy wax after passing through a filtering system in the reactor.
The high-temperature oil gas from the top of the reactor enters a circulating heat exchange separator to be subjected to heat exchange and cooling separation with circulating gas (F-T purified synthetic gas from a purification device, circulating gas from an outlet of a circulating gas compressor, hydrogen of a PSA unit and the synthetic gas after the circulating gas of a tail gas decarburization unit are mixed), and liquid phase heavy oil is sent to a stripping tower for stripping after being heated by a heavy oil stabilizing heat exchanger and a heavy oil heater. The gas phase enters an oil-gas air cooler for cooling, and then enters a light oil separator for gas-liquid separation. The separated gas phase is sent to a liquid separating tank of the circulating gas compressor and then enters the circulating gas compressor. The liquid phase separated by the light oil separator enters an oil-water separator to carry out three-phase separation of oil, water and released gas. The separated light oil is sent to an intermediate tank area; the separated release gas enters a release gas compressor; the separated synthetic water is pumped up by a synthetic water pump and is sent into an intermediate tank area. The pressure control of the whole slurry bed reaction system is placed on the pipeline from the tail gas decarburization unit to the cryogenic separation unit. The pressure of the reaction system in the start-up stage is controlled by a control valve arranged on the pipeline from the tail gas to the torch system.
After being boosted by the circulating gas compressor, the circulating gas sent to the circulating gas compressor is mixed with F-T purified synthetic gas of the self-purification device, hydrogen from the PSA unit and circulating gas from the tail gas decarbonization unit, and then enters a circulating heat exchange separator to exchange heat with high-temperature oil gas from the reactor and returns to the reactor; and the other circulating gas is continuously pressurized by a circulating gas compressor, a part of the circulating gas is sent to the tail gas decarburization unit, a part of the circulating gas is sent to a back blowing heater, the back blowing gas at the outlet of the back blowing heater enters a back blowing buffer tank to be used as the back blowing gas of the paraffin removal filter inside the reactor, and the filter is back blown under the control of a paraffin removal sequential control program.
The process of discharging heavy wax from the internal filtration system of the reactor is entirely controlled by a programmed program. The heavy wax discharged to the heavy wax collecting tank is subjected to gas-liquid separation in the collecting tank, and the separated heavy wax is sent to a stripping tower for further treatment; the gas phase parts of the 2 heavy wax collecting tanks are communicated with a heavy wax pressure stabilizing tank, and the heavy wax pressure stabilizing tank is provided with pressure control measures for controlling the pressure of the collecting tanks, so that the pressure difference and the flow of the reactor wax removal are ensured. The waxy release gas at the outlet of the heavy wax surge tank is sent to the lower part of the stripping tower.
The reactor is an isothermal reactor, the F-T synthesis reaction is a strong exothermic reaction, the reaction heat is removed by a reactor steam-water system, and the temperature of the reactor is controlled by the pressure of a steam drum. The steam-water system of the reactor comprises an upper section steam drum, an upper section steam drum circulating water pump, a lower section steam drum circulating water pump, a reactor internal heat exchanger and other devices.
The medium-pressure deoxidized water sent by the deoxidized water system is directly sent to a steam drum of the reactor, deoxidized water in the upper and lower steam drums of the reactor is respectively sent to the internal heat exchangers of the upper and lower sections of the reactor through the circulating water pumps of the upper and lower sections of the steam drums, the steam-liquid mixture after the heat taking part vaporization is returned to the steam drum for steam-liquid separation, and the separated steam is sent to a 2.8MPaG pipe network.
The pressure control valve of the steam drum is arranged on the steam manifold of the steam drum, and the pressure control point of the steam drum is determined according to the temperature of the reactor; the drum pressure in the driving stage is controlled by a pressure control valve on the blow-down line; the level control of the drum is a typical three-variable control.
Catalyst reduction unit
Raw material catalyst required by the catalyst reduction unit is conveyed to the unit by a catalyst tank truck, conveyed to a catalyst storage bin by low-pressure nitrogen gas in a pneumatic mode, conveyed to a catalyst feeding tank by a dense phase pump in a dense phase mode, and conveyed to the reduction reactor after the catalyst feeding tank is pressurized to be balanced with the pressure in the reduction reactor.
Under certain temperature and pressure, the catalyst undergoes catalyst reduction reaction in the reduction reactor under the action of the synthesis gas, and simultaneously undergoes Fischer-Tropsch synthesis reaction to produce oil products, water, carbon dioxide and other hydrocarbon compounds. Gas generated by the reduction reaction enters a heavy oil separator for gas-liquid two-phase separation after being subjected to heat exchange and cooling by a primary heat exchanger and circulating gas from a circulating compressor, the separated heavy oil goes to a heavy oil heater of a Fischer-Tropsch synthesis unit, gas phase components from the top of the separator enter a light oil separator after being further cooled by a circulating gas air cooler, liquid phase products from the bottom of the light oil separator go to the light oil water separator of the Fischer-Tropsch synthesis unit, a part of gas phase from the top of the light oil separator is mixed with purified synthetic gas from a purifying device and then is boosted by the circulating compressor, then is mixed with hydrogen from a tail gas treatment device and then enters the primary heat exchanger for heat exchange, and then enters a reduction reactor for reaction with a catalyst after being heated by a steam heater. And the other part of tail gas is sent to an FT synthesis unit. The catalyst slurry after completion of the reaction in the reduction reactor is sent under pressure to a Fischer-Tropsch synthesis unit. The reaction heat generated in the reduction reaction is taken away by deoxidized water in a steam drum, and 0.5MPa (G) low-pressure steam is generated to enter a low-pressure steam pipe network.
The heavy firewood feeding part enters a heavy firewood buffer tank of the unit from the oil product processing device, heavy firewood is driven to carry out reduction reaction, and the heavy firewood is conveyed to a reduction reactor by a heavy firewood feeding pump; the heavy firewood which is needed to be fed into the reduction reactor in the reduction reaction process is heated by a heavy firewood feeding pump through a heavy firewood heater and then is continuously sent into the reduction reactor so as to maintain the stable liquid level of the reduction reactor.
The process control data mainly involved include: feed gas component analysis data, incoming gas component analysis data, recycle gas component analysis data, decarbonizing component analysis data, feed gas flow, incoming gas flow, recycle gas component flow, decarbonizing flow, low-grade total sulfur, recycle gas compressor recycle gas component indication from a recycle gas compressor, incoming gas flow, decarbonizing purification gas flow from a tail gas decarbonizing unit, medium-temperature synthesis water flow, outlet light oil flow, heavy wax pipeline flow, steam drum outlet steam flow, outlet steam flow indication, outlet recycle water flow, self-polishing desulfurization unit Fischer-Tropsch purification gas flow, self-tail gas conversion device PSA unit hydrogen flow, outlet heavy oil flow, recycle gas compressor turbine steam inlet flow, back blowing gas flow of a tail gas decarbonizing unit, deoxidizing water flow indication adjustment, self-polishing desulfurization unit Fischer-Tropsch purification gas flow indication heavy oil liquid level indication, light oil liquid level indication, heavy wax liquid level indication, inlet high-temperature oil gas and circulating gas differential pressure indication, upper and lower low-temperature oil gas differential pressure indication of an inner foam remover, fischer-Tropsch purification air pressure indication of a self-refined desulfurization unit, decarbonization purification air pressure indication of a tail gas decarbonization unit, hydrogen pressure indication of a PSA unit of a tail gas conversion device, light oil pressure indication, inlet circulating air pressure indication, cold side differential pressure, outlet circulating air pressure indication, top low-temperature oil gas pressure indication, top release air pressure indication, wax/oil/gas all temperature indication, reactor bottom waste catalyst slurry all temperature indication, reactor heavy wax temperature indication, catalyst slurry temperature indication, reactor top high-temperature oil gas temperature indication, fischer-Tropsch purification air temperature indication of a self-refined desulfurization unit, decarbonization purification air temperature indication of a tail gas decarbonization unit, the method comprises the steps of indicating the hydrogen temperature of a PSA unit of a self-tail gas conversion device, indicating the outlet temperature of a reactor, indicating the outlet temperature of a reaction outlet cooler, indicating the circulating gas component of a self-circulating gas compressor, indicating the circulating gas component of the self-circulating gas compressor, indicating a silica analyzer, a conductivity detector, the flow rate of an incoming tower, the decarburization purifying gas flow rate of a tail gas decarburization unit, indicating the hydrogen flow rate of the PSA unit of the self-tail gas conversion device, indicating the flow rate of medium-temperature synthetic water, stabilizing heavy wax to a stripping tower, indicating the flow rate of a heavy wax pipeline, indicating the flow rate of the heavy wax pipeline, indicating the flow rate of turbine steam inlet of the circulating gas compressor, indicating and adjusting the back-blowing gas flow rate of a desquamation decarburization unit, and adjusting the back-blowing gas flow rate of the desquamation unit, and indicating the oil-water boundary position.
S2, establishing a mechanism model and a process model by using the sample data in the S1; the process model is established for the Fischer-Tropsch synthesis reaction process according to the Fischer-Tropsch synthesis reaction design file, the DCS picture and the operation habit, and the operation process and the related data of the reaction process are simulated.
S3, carrying out data characteristic analysis by using the output data of the mechanism model, the output data of the process model and the data samples in S1 established in the S2, and finding out the relation between different types of data; the data characteristic analysis comprises the following analysis methods:
correlation analysis, namely, finding the size and positive-negative correlation of correlation coefficients among different types of data; and carrying out correlation analysis on the preprocessed and standardized data, and according to the principle that one variable among two variables with stronger correlation is selected as a characteristic variable, combining expert experience assistance to realize the primary selection of the characteristic variable, wherein the correlation analysis flow is shown in figure 2.
Principal component analysis, finding out main influencing factors of data (key data to be referred to in actual business); the principal component algorithm is used to extract the feature variables, in this embodiment, the first 4 principal components are used to determine that the accumulated variance contribution of the principal components is greater than 85%, and each principal component is a linear combination of other indexes, rather than a specific certain index, and the principal component analysis flow is shown in fig. 3.
Causal link analysis, namely, finding causal relation of influencing factors of data (key data to be referred to in actual business); the acquisition device comprises a full amount of historical data including data of process, quality, equipment, raw materials and the like, performs pretreatment such as setting, filtering, standardization and the like on the data, then calculates the transfer entropy of variables related to model output variables by using a transfer entropy algorithm, draws a causal link diagram of the model output variables based on the calculation result of the transfer entropy, and then screens characteristic variables by combining expert experience from root causes. The causal link analysis flow is shown in fig. 4.
And (3) an expert experience analysis method, which integrates field expert experiences and finds out an empirical optimization rule boundary.
S4, dividing working conditions by utilizing the data features in the S3, and constructing and outputting a data model; drawing dynamic health datum lines applicable to different working conditions and corresponding index health value ranges according to the data model simulation; in the health parameter range, the optimal operation parameters and schemes corresponding to the optimal performance indexes under different working conditions are found through an optimizing model, and the following model is adopted for constructing the data model: 1. the neural network is used for automatically identifying various working conditions and pushing an optimal operation scheme. 2. The random forest and the decision tree are used for steady-state operation schemes and pushing under different working conditions. 3. And multiple linear regression, which is used for operation indexes and a prediction model of results and performance indexes. The support vector machine method is used for a system auxiliary decision-making model.
And outputting the analysis data model; the analytical data model includes the following:
predictive analysis model, based on the prediction of the service life of the catalyst by machine learning, based on indexes such as CO conversion rate, dry gas, light oil, heavy wax selectivity, oil-water ratio, using time, catalyst stock, unit consumption and the like as target indexes, constructing a regression model, constructing a machine learning mechanism and the like, constructing a regression model, constructing a machine learning mechanism, and adopting a supervised algorithm to realize the prediction of the residual service life of the catalyst; the method effectively guides the production scheduling, catalyst purchasing and catalyst replacement business development;
the steady-state operation model is used for constructing an advanced early-warning model based on the health early-warning management applied by the advanced early-warning model, and selecting data of normal working conditions of the catalyst to realize the prediction of health values; based on an approximation degree evaluation algorithm, evaluating the difference degree between an actual value and a predicted value, and realizing early warning of index abnormality; the purposes of interfering the process operation in advance and ensuring the long-term healthy operation of the catalyst are achieved; the flow chart of the running health index analysis is shown in fig. 5.
The collaborative optimization model is constructed based on an optimal process sample library of a partitioning algorithm, the partitioning algorithm is adopted to perform on-site working condition clustering, the working conditions are partitioned, a working condition identification model is built according to working condition partitioning results, and optimal process parameters of each working condition are calculated; collecting production data in real time, identifying working conditions, finding out optimal technological parameters under the working conditions, recommending the optimal technological parameters to an operator for production guidance, and enabling a stable optimization flow diagram based on health values to be shown in fig. 6;
The performance management model combines the self-defined health threshold value and the optimization range threshold value of the service requirement through data acquisition of each module of the system, combines the management of the thresholds to form performance evaluation indexes of practice and operation processes, ranks the performance of each management object based on the performance evaluation indexes, comprehensively carries out performance evaluation and management according to different time granularity, and forms a closed-loop management model of services such as perception, alarm, intervention, health guarantee, working condition optimizing, performance evaluation, prediction and the like. And (3) drawing a dynamic health datum line suitable for different working conditions according to the data model simulation.
S5, evaluating the practicability, applicability, accuracy, errors and effectiveness of the data model obtained in the S4 and the feasibility of the model, and constructing and training the model by using 80% of the sample data in the S1, wherein 20% of the data are used for verification and evaluation; the evaluation method adopts a model accuracy method, a root mean square error method, a root mean square logarithmic error method and a relative error method; after being qualified, the data model is deployed as a functional module component, and the unqualified data model is returned to S1 for processing due to the problem of data samples; returning the unqualified data model to S2 for processing due to the mechanism model problem; returning the unqualified data model to S3 for processing due to the characteristic analysis problem; in this embodiment, the test is determined to be acceptable when the accuracy is greater than 95% and the relative error is less than 3% both satisfy the set threshold.
And when the accuracy of the model meets the service requirement, model deployment can be performed. After the model is deployed, online data is accessed to perform stream calculation, and online calculation and display of indexes such as product quality can be realized. The model deployment flow is shown in fig. 7.
S6, collecting real-time operation data in the Fischer-Tropsch synthesis reaction, accessing the functional module established in the S5, and comparing the real-time operation data with the output result of a data model in the functional module; the management model is shown in fig. 8;
1. and for the prediction module, after the indexes are compared, the change trend is presented, the change direction is qualitatively judged, and the future change trend and range are quantitatively presented.
2. And comparing the steady-state operation modules, and carrying out early warning on the real-time value and the threshold value of the output value of the model, and providing an adjustment suggestion or an index deviation correction suggestion to ensure that the index operates in a stable and healthy range.
3. And comparing the collaborative optimization models, determining whether optimization is needed under the working condition, and providing a parameter optimization scheme and an index control scheme within a stable and healthy range if the performance index is cracked and deviates from a threshold value.
In this embodiment, the algorithm model adopts a visual design, and the visual interface is shown in fig. 9. The model of the present embodiment has the following functions:
(1) Single objective optimization analysis
The single-target optimization analysis refers to searching an optimal value of a target under a certain raw material condition and a corresponding strongly-correlated operation variable in an operation sample library. The realization process comprises the following four steps: clustering and analyzing raw materials; establishing a classification model; forming an operation sample library; and (5) optimizing parameters.
1.1 Cluster analysis of working conditions
And carrying out cluster analysis on main parameters of the device, and accurately calculating the working conditions of the device.
1.2 building of Classification model
And establishing a classification model according to the clustering result, and realizing online identification of the current working condition.
1.3 creation of operation sample library
And writing the key process parameters, the types of working conditions, the values of the optimization targets, the strong correlation adjustable parameters and the like into a writing operation sample library.
1.4 parameter optimization
Accessing online data, judging working conditions, calculating the corresponding value of the strong related operation variable when the historical target parameter is optimal under the current working conditions, and displaying
(2) Multi-objective optimization analysis
2.1 Cluster analysis of working conditions
And carrying out cluster analysis on main parameters of the device, and accurately calculating the working conditions of the device.
2.2 building of Classification model
And establishing a working condition identification model according to the result of the cluster analysis, accessing online data, and judging the current working condition in real time.
2.3 creation of operation sample library
Writing the steady state values of key process parameters, the types of working conditions, the steady state values of optimization targets, the steady state values of the strongly-related adjustable parameters and the like into a write operation sample library.
2.4 determination of theoretical optimum point
Under the condition of searching a certain type of raw materials in the operation sample library, the optimal values of a plurality of selected optimization variables are used as the coordinates of theoretical optimal points in a multidimensional space in a certain load interval. The number of the optimization variables is selected to be the coordinate dimension of the theoretical optimal point. Such as: the user selects 4 optimization variables, and then the coordinate of the theoretical optimal point is 4 dimensions, and the value of the coordinate is the optimal value of the 4 optimization variables under the condition of a certain type of raw materials.
After the user selects the optimization variables, the coordinates of the theoretical optimum point are determined according to the operation in the previous step. And under the working condition, taking the value of the selected optimization variable in the operation sample library as the coordinate of the sample point. And calculating the Euclidean distance between the sample point and the theoretical optimal point. After the calculation is completed, the Euclidean distances are sorted from small to large. And selecting a sample point with the smallest Euclidean distance as the optimization result of the time. The optimization result comprises the optimal value of the selected optimization variable and the value of the corresponding strong correlation variable.
While the fundamental and principal features of the invention and advantages of the invention have been shown and described, it will be apparent to those skilled in the art that the invention is not limited to the details of the foregoing exemplary embodiments, but may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present disclosure describes embodiments, not every embodiment is provided with a separate embodiment, and that this description is provided for clarity only, and that the disclosure is not limited to the embodiments described in detail below, and that the embodiments described in the examples may be combined as appropriate to form other embodiments that will be apparent to those skilled in the art.
Claims (10)
1. An intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst is characterized in that: the method comprises the following steps:
collecting DCS, SCADA, PLC in the Fischer-Tropsch synthesis reaction process and historical operation data of an application system, wherein the historical operation data comprise working condition parameters, working condition parameters and target data, the working condition parameters comprise raw material gas composition, raw material gas H/C, raw material gas sulfur content, raw material gas feeding amount and raw material gas hydrogen distribution amount, and carrying out data treatment on the data to obtain treated data samples;
step two, a mechanism model and a process model are built by using the sample data in the step one, and the process model is built according to a Fischer-Tropsch synthesis reaction design file, a DCS picture and operation habits, so that the operation process and related data of the reaction process are simulated;
thirdly, carrying out data characteristic analysis by utilizing the output data of the mechanism model, the output data of the process model and the data sample in the first step, and finding out the relation between different types of data;
dividing working conditions by utilizing data characteristics in the third step, wherein the flow rate of raw materials at the inlet of the reactor and CO in the crude synthesis gas in the third step 2 、H 2 S content, CH 4 Raw material gas H/C and raw material gas hydrogen preparationAfter combining different intermediate data of the quantity data, forming a plurality of working condition partitions, and constructing a data model based on different operation data sets in the working condition partitions;
step five, evaluating the practicability, applicability, accuracy, errors and effectiveness of the data model obtained in the step four and the feasibility of the model, deploying the data model as a functional module assembly after the data model is qualified, and returning the unqualified data model to the step one for processing due to the problem of data samples; returning the unqualified data model to the second step for treatment due to the mechanism model problem; returning the unqualified data model to the third step for processing due to the characteristic analysis problem;
step six, collecting real-time operation data in the reaction process of preparing olefin from methanol, accessing the functional module established in the step five, and comparing the data with the output result of a data model in the functional module; pushing an optimal operation control scheme under the current working condition when the result index or the performance index is degraded according to the comparison result pushing operation suggestion; and when the result index or the performance index is kept stable, namely the deviation from the output result value is within a certain range, the current situation is kept, and the index optimization operation is not performed.
2. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 1, wherein the intelligent operation management method comprises the following steps: the operation data in the first step comprises raw gas component analysis data, tower gas component analysis data, circulating gas component analysis data, decarbonizing component analysis data, raw gas flow, tower gas flow, circulating gas flow, decarbonizing flow, low-alpha tail gas total sulfur, circulating gas component of a self-circulating gas compressor, tower gas flow, decarbonizing and purifying gas flow of a tail gas decarbonizing unit, medium-temperature synthetic water flow, outlet light oil flow, heavy wax pipeline flow, steam drum outlet steam flow, outlet steam flow indication, outlet circulating water flow, fischer-Tropsch and purifying gas flow of a self-refining desulfurization unit, hydrogen gas flow of a tail gas conversion device PSA unit, outlet heavy oil flow, turbine steam inlet flow of a circulating gas compressor, back blowing and deoxidizing gas flow of a tail gas decarbonizing unit Fischer-Tropsch gas flow rate of the self-refined desulfurization unit, heavy oil liquid level, light oil water level, heavy wax liquid level, inlet high-temperature oil gas and circulating gas differential pressure, upper and lower low-temperature oil gas differential pressure of an inner demister, fischer-Tropsch gas pressure of the self-refined desulfurization unit, decarbonizing gas pressure of the self-tail gas decarbonizing unit, pressure of hydrogen of a PSA unit of a self-tail gas conversion device, pressure of light oil, inlet circulating gas pressure, cold side differential pressure, outlet circulating gas pressure, top low-temperature oil gas pressure, top gas release pressure, wax/oil/gas all temperatures, all temperatures of waste catalyst slurry at the bottom of a reactor, reactor heavy wax temperature, catalyst slurry temperature, top high-temperature oil gas temperature of the reactor, fischer-Tropsch gas temperature of the self-refined desulfurization unit, decarbonizing gas temperature of the self-tail gas decarbonizing unit, pressure of the PSA unit of the self-tail gas conversion device, temperature deviation of the outlet of the reactor, and the like, the method comprises the steps of reactor outlet cooler outlet temperature, self-circulating gas compressor circulating gas component, silicon dioxide, conductivity, tower inlet gas flow, decarbonizing and purifying gas flow from a tail gas decarbonizing unit, hydrogen flow from a tail gas converting device PSA unit, medium-temperature synthetic water flow indication, stabilizing heavy wax to a stripping tower, heavy wax pipeline flow, circulating gas compressor turbine steam inlet flow, back blowing gas flow of a tail gas decarbonizing unit and oil-water boundary position.
3. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 2, wherein the intelligent operation management method comprises the following steps: the target data comprise light oil yield, heavy wax yield, synthetic water yield, CO conversion rate, selectivity of each product, catalyst activity index, product yield, catalyst consumption, unit energy consumption and unit consumption of raw material gas.
4. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 1, wherein the intelligent operation management method comprises the following steps: the data management in the first step comprises one or more of the following management methods: data cleaning, data filtering, data setting, data standardization and data alignment.
5. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 4, wherein the intelligent operation management method comprises the following steps: the data characteristic analysis in the third step comprises one or more of the following analysis methods:
correlation analysis, namely, finding the size and positive-negative correlation of correlation coefficients among different types of data;
principal component analysis, finding out main influencing factors of data;
causal link analysis, namely, finding causal relation of influencing factors of data;
and (3) an expert experience analysis method, which integrates field expert experiences and finds out an empirical optimization rule boundary.
6. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 1, wherein the intelligent operation management method comprises the following steps: the data model includes.
7. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 1, wherein the intelligent operation management method comprises the following steps: in the fourth step, dynamic health datum lines applicable to different working conditions and corresponding index health value ranges are simulated and drawn according to the data model; in the health parameter range, finding out the optimal operation parameters and schemes corresponding to the optimal performance indexes under different working conditions through an optimizing model; the data model comprises one or more of a predictive analysis model, a steady-state operation model, a collaborative optimization model and a performance management model.
8. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 7, wherein the intelligent operation management method comprises the following steps: the predictive analysis model is used for predicting the service life of the catalyst based on machine learning, and is used for constructing a regression model and establishing a machine learning mechanism based on indexes such as CO conversion rate, dry gas, light oil, heavy wax selectivity, oil-water ratio, use duration, catalyst inventory, unit consumption and the like as target indexes, and predicting the residual service life of the catalyst by adopting a supervised algorithm;
The steady-state operation model is used for constructing an advanced early-warning model based on the health early-warning management applied by the advanced early-warning model, and selecting data of normal working conditions of the catalyst to realize the prediction of health values; based on an approximation degree evaluation algorithm, evaluating the difference degree between an actual value and a predicted value, and realizing early warning of index abnormality; the purposes of interfering the process operation in advance and ensuring the long-term healthy operation of the catalyst are achieved;
the collaborative optimization model is constructed based on an optimal process sample library of a partitioning algorithm, the partitioning algorithm is adopted to perform on-site working condition clustering, the working conditions are partitioned, a working condition identification model is built according to working condition partitioning results, and optimal process parameters of each working condition are calculated;
the performance management model combines the self-defined health threshold value and the optimization range threshold value of the service requirement through data acquisition of each module of the system, combines the management of the thresholds to form performance evaluation indexes of practice and operation processes, ranks the performance of each management object based on the performance evaluation indexes, comprehensively carries out performance evaluation and management according to different time granularity, and forms a closed-loop management model of services such as perception, alarm, intervention, health guarantee, working condition optimizing, performance evaluation, prediction and the like.
9. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 8, wherein the intelligent operation management method comprises the following steps: the construction method of the data model comprises one or more of the following methods: neural network method, support vector machine method, decision tree method, multiple linear regression method, and random forest.
10. The intelligent operation management method for the Fischer-Tropsch synthesis slurry bed catalyst according to claim 7, wherein the intelligent operation management method comprises the following steps: the evaluation method in the fifth step comprises one or more of the following methods: model accuracy method, root mean square error method, root mean square logarithmic error method, and relative error method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311595841.7A CN117612619A (en) | 2023-11-28 | 2023-11-28 | Intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311595841.7A CN117612619A (en) | 2023-11-28 | 2023-11-28 | Intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117612619A true CN117612619A (en) | 2024-02-27 |
Family
ID=89951019
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311595841.7A Pending CN117612619A (en) | 2023-11-28 | 2023-11-28 | Intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117612619A (en) |
-
2023
- 2023-11-28 CN CN202311595841.7A patent/CN117612619A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AlNouss et al. | A techno-economic-environmental study evaluating the potential of oxygen-steam biomass gasification for the generation of value-added products | |
| CN109814513B (en) | Catalytic cracking unit optimization method based on data model | |
| Zhang et al. | Effective MILP model for oil refinery-wide production planning and better energy utilization | |
| Sharma et al. | Multi‐objective optimization applications in chemical engineering | |
| CN103119130A (en) | Active cloud point controller for refining applications and related method | |
| CN113724800A (en) | Catalytic cracking unit simulation prediction method based on molecular level mechanism model and big data technology | |
| CN106709654A (en) | Global operating condition evaluating and quality tracing method for hydrocracking process | |
| CN108647373A (en) | A kind of industrial process flexible measurement method based on xgboost models | |
| CN101187803B (en) | Ammonia converter production optimization method based on data excavation technology | |
| Penteado et al. | Techno-economic evaluation of a biogas-based oxidative coupling of methane process for ethylene production | |
| CN110456756A (en) | A method of suitable for continuous production process overall situation operation conditions online evaluation | |
| CN115188429A (en) | Catalytic cracking unit key index modeling method integrating time sequence feature extraction | |
| CN110376987A (en) | The oil refining process method for on-line optimization and device detected based on case library and stable state | |
| CN103087750A (en) | Naphtha industrial cracking furnace value maximization model construction method | |
| Poto et al. | Techno-economic assessment of the one-step CO2 conversion to dimethyl ether in a membrane-assisted process | |
| CN108287474B (en) | Robust operation optimization method of catalytic reforming reactor based on uncertainty of raw materials | |
| KR102489086B1 (en) | Knowledge-based assessment flatform for carbon dioxide direct hydrogenation catalysts | |
| CN117612619A (en) | Intelligent operation management method for Fischer-Tropsch synthesis slurry bed catalyst | |
| Guerra et al. | Green propylene and polypropylene production from glycerol: process simulation and economic evaluation | |
| CN110021377B (en) | Method and device for predicting deactivation of hydrocracking catalyst and storage equipment | |
| CN103087753A (en) | Heavy raw material industrial cracking furnace value maximization model construction method | |
| CN117672392A (en) | Intelligent operation management method for methanol synthesis catalyst | |
| CN117542429A (en) | Hydrocracking fixed bed catalyst management method | |
| CN101145029A (en) | P-xylene oxidation reaction operation condition optimization method | |
| CN111598305B (en) | Light hydrocarbon separation device running state optimization and prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |